This study used a randomized, double-blind, placebo-controlled, crossover design with a 14-day washout period between experimental sessions. Participants visited the laboratory on three occasions. In the familiarization visit, participants performed the 30-minute treadmill time trial (TT) in normobaric hypoxia (FiO2=15.5 %) and a graded running test to volitional exhaustion in normoxia to determine maximal oxygen consumption (VO2max). The last two visits served as experimental sessions. During each experimental session, runners completed a 30-minute TT in normobaric hypoxia. Heart rate (HR), rating of perceived exertion (RPE), and oxygen saturation (SaO2) were monitored throughout the TT. Blood lactate was assessed prior to, as well as 5, 15 and 30 min following the TT. Additionally, saliva and fraction of exhaled nitric oxide (FENO) were sampled pre- and post-TT. In the four days before each experimental session, participants supplemented with 4 days of blueberry powder (BLU) or 4 days of a placebo powder (PLA). For each participant, sessions were conducted at the same time of day (± 30 min).
Eleven recreational runners (4 males; 7 females) volunteered for this study (mean ± SD: age = 28.4 ± 7.5 years; height = 170.9 ± 9.0 cm; body mass = 65.6 ± 9.7 kg; VO2max = 49.9 ± 9.3 mL·kg− 1·min− 1). All participants were non-smokers; free from cardiovascular, respiratory, and oral/gum disease; and free from musculoskeletal injury. None of the participants had been exposed to altitude (natural or simulated equivalent to ≥ 1200 m above sea level) within the 3 months prior to the start of the study.
Graded running test and VO2max
During the familiarization session, the graded running test began at a treadmill speed considered comfortable by the participant (Alpine Runner, True Fitness Technology, St. Louis, MO). Treadmill gradient was increased by 2 % every 2 min for the first 6 min of the test and then by 1 % every 1 min until volitional exhaustion. Oxygen consumption was measured continuously throughout the test using a gas analysis system (TrueOne, Parvomedics, Sandy, Utah). VO2max was determined as the highest 30-s average. All participants performed a 5-minute warm-up at a self-selected speed before commencing the graded running test.
Prior to the start of each experimental session, participants rinsed their mouth with distilled water and were fitted with a HR sensor and chest strap (Polar H7, Kempele, Finland). Participants were then moved into the altitude chamber (Altitudetech Inc, Kingston, ON, Canada) where they rested in a seated position for 10 min. During this time each participant’s food/activity log was reviewed to ensure compliance with study requirements.
Afterwards, participants provided saliva samples and pre-TT measures of FENO and blood lactate were administered. Blood lactate was determined using a finger prick and handheld lactate analyzer (Lactate Pro; Arkray KDK Corp., Kyoto, Japan). The Lactate Pro analyzer, when compared to laboratory-based analyzers, has been shown to be both reliable (Coefficient of variation [CV] ranging from 2.7 to 5.7 %) and valid (SEE ranging from 0.5 to 1.1; Bias raging from − 0.6 – -1.7) [16,17,18]. FENO was assessed using a handheld electrochemical analyzer (NIOX MINO, Aerocrine AB, Solna, Sweden) following previously established guidelines. Once the pre-TT measures were completed, participants completed a 5-minute self-paced warm-up followed by the 30-minute TT; both of which were performed on a non-motorized treadmill (Trueform Runner; Model: Performance; Connecticut, USA). Runners were instructed to run as far as possible during the 30 min (i.e. best effort). Runners were blinded to distance and running speed but were aware of the elapsed time. A fan was placed in front of the participant to simulate outdoor running conditions. A non-motorized treadmill was utilized for the TT because changes in running speed (i.e. pacing) are instantaneous, influenced by subconscious fatigue, and characteristic of over-ground running. In experienced runners, like those in the present study, a TT on non-motorized treadmill has been shown to be reliable after a single familiarization session. HR and SaO2, by fingertip pulse oximeter (MightySat, Masimo Corp, Irvine, CA), were recorded throughout the TT while RPE (Borg 6–20 scale) was recorded every 5 min.
Upon completion of the 30-minute TT, saliva was sampled, FENO was assessed at 4 min post-TT, and blood lactate measurements were repeated at 5, 15, and 30 min post-TT. To assess lactate recovery, all post-TT values were normalized to the 5-minute post-TT value.
Saliva was collected to measure cortisol, uric acid (UA), interleukin-6 (IL-6), and C-reactive protein (CRP). Salivary biomarkers were measured due to the ease and non-invasive nature of sample collection. The biomarkers analyzed in the present study have been shown to be valid and reliable indicators of the physiological stress to running[22, 23]. At each of the two saliva collection time points, two ml of saliva were collected using the passive drool method. Following saliva collection, samples were frozen at -20 °C until analyses were conducted. Saliva samples were assayed using commercially available highly sensitive enzyme linked immunoassay kits (Salimetrics, PA, USA). Samples were analyzed in duplicate. The intra-assay CVs for the above-mentioned saliva assays were below 5 %.
Supplementation and Diet/Physical Activity Restrictions
In the four days leading up to each experimental session, runners ingested either freeze-dried blueberry powder (BLU) (50/50 blend of Vaccinium virgatum/Vaccinium corymbosum, Oxygen Radical Absorbance Capacity [ORAC] = 831 µmole TE/g, Table 1) or placebo powder (PLA). PLA, manufactured to have a similar taste and appearance to the blueberry powder, was made from maltodextrin, fructose, dextrose, citric acid, malic acid, silicon dioxide (flow agent), xanthan gum, artificial colors, and artificial and natural flavours.
During the supplementation period, participants ingested 3, 24 g packets of blueberry or placebo powder per day (BLU = ~ 500 g of fresh blueberries/day). Ingestion of the packets was distributed evenly throughout the day with the final packet ingested 2 h prior to the start of the session. For ingestion, the powder was mixed with about 250 ml of water. The antioxidant and anthocyanin content of the blueberry powder used in the present study (Table 1) was assessed in the batch by a commercial laboratory (International Chemistry Testing, Milford, MA). Antioxidant activity was assessed using the Oxygen Radical Absorbance Capacity (ORAC) test based on the methods of Ou et al.. The anthocyanin assay was based on the methods described by Lee et al..
During each 4-day supplementation period participants were asked to limit the intake of other foods/drinks high in polyphenols as well as refrain from taking other supplements and anti-inflammatory medications. Additionally, participants avoided caffeine and alcohol intake 8 and 24 h (respectively) prior to the start of each session. Participants also abstained from moderate to vigorous intensity physical activity within 24 h of each session. To facilitate this, participants completed a 4-day food and physical activity diary before the first experimental session and were instructed to replicate this diary before the second experimental session.
Normality of the data was evaluated using the Shapiro–Wilk test. Mauchly test of sphericity was performed to assess the homogeneity of data. Where violations were present, Greenhouse–Geisser adjustments were made. Between-supplement differences in 30-min running distance, average HR, maximum HR, average SaO2, RPE, pre-TT salivary biomarkers, and pre- to post-TT change in salivary biomarkers were analyzed using a paired t-test. Differences in FENO and blood lactate were analyzed using a repeated measures ANOVA with 2 supplement conditions by 2 time points (1 pre- and 1 post-TT measure). Differences in blood lactate recovery were analyzed using a repeated measures ANOVA with 2 supplement conditions by 3 time points (5-min, 15-min, and 30-min post-TT measures). Post hoc analysis was performed using t-tests with a Bonferroni correction. Statistical significance was set at p < 0.05. To estimate the magnitude of the effects from supplementation, Cohen’s d effect sizes (ES) were calculated with the magnitude of effects considered small (0.20–0.49), moderate (0.50–0.79) and large (> 0.80). Additionally, Pearson correlation coefficients were used to explore relationships between the study variables. Data are presented as mean ± SD. All statistical analyses were conducted using SPSS v 20.0 (SPSS Inc., Chicago, IL, USA).